DOI:
10.1039/C6RA03872G
(Paper)
RSC Adv., 2016,
6, 37246-37253
Diamide-based fullerosteroidal and disteroidal [2]rotaxanes: solvent-induced macrocycle translocation and/or unthreading†
Received
11th February 2016
, Accepted 28th March 2016
First published on 30th March 2016
Abstract
The synthesis, characterization and behaviour of two novel Leigh-type amide [2]rotaxanes are reported. NMR study shows that fullerosteroidal and disteroidal rotaxanes occur in a peptide co-conformation in CDCl3. [D6]DMSO induces fast unthreading of disteroidal rotaxane, which includes steroid co-conformers as intermediates. On the other hand, fullerosteroidal rotaxane undergoes predominantly a shuttling process occupying the stacked co-conformation, whereas unthreading is very slow in comparison with its disteroidal analogue (25% after 7 days). Moreover, organogelation and self-organization properties were studied. It was found that disteroidal rotaxane is an organogelator and its SEM image shows that it forms a branched film-like network in a PhMe/EtOAc 1
:
1 mixture.
1. Introduction
Rotaxanes are mechanically interlocked supramolecular architectures composed of one or more linear moieties (thread, axle) surrounded by macrocyclic ring(s) (wheel) and sufficiently large stopper groups positioned at the axle termini in order to prevent the dissociation process.1 Due to their abacus like structure,2 rotaxanes have attracted great attention in studies of non-covalent bonding interactions, as well as in wide ranges of applied research. Such compounds have proven to be valuable in biomedical exploration (enzyme-sensitive rotaxane-based propeptides,3 supramolecular nanovalves as anticancer drug carriers,4 rotaxanes with oligomeric axles as intracellular transport agents,5 squaraine-rotaxanes as molecular probes for in vitro and in vivo fluorescence cell imaging6), but also as components of nanoelectrochemical devices.7 Besides careful design based on a choice of axle–wheel interactions (hydrophobic, ionic, π–π stacking, hydrogen bonding, metal–ligand coordination) and consequent rational subunit selection, the development of template-directed synthetic strategies1a,8 expanded the possibility to obtain a large number of rotaxanes. The notable mobility of the macrocyclic ring within the supramolecular system enables its rotation around the thread1a and translational movement along the axle allows reversible (shuttling)9 or irreversible (unthreading) movement.10 In addition, the shuttling process in rotaxanes possessing two or more different stationary units causes co-conformational isomerism,11 and the predominance of one of the translational isomers in response to external stimuli. The position of the ring component of the molecular shuttle can be controlled by variation of the electrochemical potential,12 pH,13 illumination,14 solvent,9 or by cooperative effects of various driving forces15 which allow the manipulation of binding affinities. Efficient switching without further chemical transformations can be performed using solvents of different polarity (dichloromethane and dimethyl sulfoxide), as described by the examples of two hydrogen bond-assembled fullerene-stoppered rotaxanes.2,16 However, poor solubility of Leigh-type fullerene-stoppered rotaxanes is one of the major obstacles to their application.2,17 It was also observed that the presence of the steroidal subunit in mechanically interlocked molecules played an essential role in their self-organization to more complex, morphologically well defined architectures.18
Willing to examine individual as well as mutual effects of steroidal and fullerenic stoppers on properties of mechanically interlocked systems, here we describe the template-directed synthesis and detailed characterization of two novel hydrogen bond assembled rotaxanes – disteroidal, with pregnenolone-derived stoppers, and fullerene-pregnenolone stoppered. Their shuttling/unthreading behaviour in solvents of different polarity as well as the self-organization process are also examined and discussed.
2. Experimental
General information
The amine-TFA salt 1,19 fulleropyrrolidinic acid 2a,20 pregnenolone hemisuccinate 2b21 and the thread 3a19 were synthesized according to the literature procedures. Chloroform and triethylamine (TEA) were dried by refluxing with P2O5 and with CaH2, respectively, and distilled prior to use. Flash column chromatography (FCC) and dry-column flash chromatography (DCFC) were carried out with Merck silica gel 0.04–0.063 mm and 0.015–0.04 mm, respectively. Thin layer chromatography (TLC) was carried out on precoated silica gel 60 F254 plates. Melting points were determined on a Digital melting point WRS-1B apparatus and are uncorrected. IR spectra (ATR) were recorded with a Perkin-Elmer-FT-IR 1725X spectrophotometer; ν values are given in cm−1. 1H and 13C NMR spectra were recorded with Varian Gemini 200 (1H at 200 MHz, 13C at 50 MHz) and Bruker Avance spectrometers (1H at 500 MHz, 13C at 125 MHz). Chemical shifts (δ) are expressed in ppm and coupling constants (J) in Hz. TMS was used as an internal reference. The homonuclear 2D (DQF-COSY) and the heteronuclear 2D 1H–13C spectra (HSQC, HMBC) were recorded with the usual settings. UV spectra were recorded with a GBC-Cintra 40 UV/Vis spectrophotometer. The high-resolution MS spectra were taken with Agilent 6210 LC ESI-MS TOF and LTQ Orbitrap XL spectrometers. Standard steroidal numbering was used together with abbreviations for 4-aminobutanoic and succinate fragments (GABA and Succ, respectively). Labels of the peptide/diamide moieties and the macrocycle ring in threads and rotaxanes are given in Scheme 2. SEM: the morphology investigations were carried out with SEM, using a JEOL JSM-840A instrument, at an acceleration voltage of 30 kV. The dried samples obtained from a drop of dilute solutions of rotaxanes 4a (0.1 mM in CHCl3) and 4b (0.1 mM in PhMe/EtOAc 1
:
1, v/v) were deposited on the surface of a Si substrate (5 × 5 mm) and left overnight to slowly evaporate in a glass Petri dish (diameter 10 cm) under a PhMe atmosphere at room temperature. The investigated samples were gold sputtered in a JFC 1100 ion sputter device and then subjected to SEM observations.
Thread 3a19. UV/Vis (CHCl3): λmax(ε) = 430 nm (2.33 × 103 mol−1 dm3 cm−1); 1H NMR (500 MHz, [D6]DMSO): δ = 8.11 (br t, J = 5.0 Hz, 1H, NH(Gly)), 7.83 (br t, J = 5.5 Hz, 1H, NH(GABA)), 5.34 (m, 1H, HC(6)), 4.49 (m, 1H, HC(3)), 4.45 (s, 4H, H2C(pyrr)), 3.78 (br d, J = 5.0 Hz, 2H, H2C(Gly)), 3.16 (q, J = 6.5 Hz, 2H, H2C(4′)), 3.13 (t, J = 7.0 Hz, 2H, H2C (4′′)), 2.54 (m, 3H, H2C(2′′) and HC(17)), 2.29 (m, 4H, H2C(2′) and H2C(4)), 2.19 (quint, J = 7.0 Hz, 2H, H2C(3′′)), 2.08 (s, 3H, H3C(21)), 2.06 (m, 1H, HC(16)), 2.02 (m, 1H, HC(12)), 1.95 (m, 1H, HC(7)), 1.85 (m, 1H, HC(1)), 1.81 (m, 1H, HC(2)), 1.72 (m, 2H, H2C(3′)), 1.64 (m, 1H, HC(15)), 1.60 (m, 1H, HC(11)), 1.58 (m, 1H, HC(16)), 1.56 (m, 1H, HC(2)), 1.55 (m, 1H, HC(7)), 1.43 (m, 5H, HC(8), HC(11) and HC(12)), 1.16 (m, 2H, HC(14) and HC(15)), 1.00 (m, 1H, HC(9)), 0.98 (s, 3H, H3C(19)), 0.55 (s, 3H, H3C(18)); 13C NMR (125 MHz, [D6]DMSO): δ = 208.2 (C(20)), 172.7 (C(1′′)), 171.9 (C(1′)), 169.1 (CO(Gly)), Cfull (155.2, 146.7, 145.9, 145.7, 145.5, 144.8, 144.7, 144.1, 142.6, 142.1, 141.8, 141.6, 141.3, 139.5, 135.8), 139.3 (C(5)), 121.9 (C(6)), 73.1 (C(3)), 70.5 (Cfull-sp3), 67.1 (C(pyrr)), 62.7 (C(17)), 56.1 (C(14)), 53.6 (C(4′′)), 49.3 (C(9)), 43.3 (C(13)), 42.3 (C(Gly)), 38.1 (C(12)), 37.9 (C(4′)) 37.7 (C(4)), 36.5 (C(1)), 36.1 (C(10)), 33.4 (C(2′′)), 31.3 (C(8)), 31.1 (C(7)), 27.4 (C(2)), 24.5 (C(3′′)), 24.3 (C(15)), 24.0 (C(3′)), 22.3 (C(16)), 20.6 (C(11)), 19.0 (C(19)), 12.9 (C(18)).
Thread 3b. To an ice bath cooled solution of the TFA salt 1 (135 mg, 0.236 mmol, 1 equiv.) in DCM (10 mL), pregnenolone hemisuccinate 2b (147 mg, 0.354 mmol, 1.5 equiv.), TEA (30 mg, 41 μL, 0.295 mmol, 1.25 equiv.), and HOBT (58 mg, 0.378 mmol, 1.6 equiv.) were added. A solution of DCC (78 mg, 0.378 mmol, 1.6 equiv.) in DCM (5 mL) was added dropwise over 2 h and the reaction mixture was stirred for 24 h. The solvent was evaporated in vacuo and the residue chromatographed by DCFC on SiO2 column using PhMe/EtOAc/MeOH 5
:
5
:
2 to obtain 3b as a colourless oil (128 mg, 63%), which was precipitated from the DCM solution with n-hexane. Rf = 0.54 (PhMe/EtOAc/MeOH 5
:
5
:
1); mp 119–120 °C; IR:
= 3305, 3079, 2941, 1729, 1707, 1650, 1553, 1440, 1358, 1253, 1177, 1025 cm−1; 1H NMR (500 MHz, CDCl3): δ = 6.79 (t, J = 6.0 Hz, 1H, NH(GABA)), 6.47 (t, J = 5.5 Hz, X part of ABX system, 1H, NH(Gly)), 5.37 (t, J = 5.0 Hz, 2H, HC(6)), 4.60 (m, 2H, HC(3)), 3.93 (two AB quartets, AB part of ABX system, J(AB) = 15 Hz, J(AX, BX) = 5.5 Hz, 2H, H2C(Gly)), 3.29 (q, J = 7.0 Hz, 2H, H2C(4)-GABA), 2.70 (dt, J = 2.0 and 6.5 Hz, 2H, H2C(3)-succ), 2.53 (two t, 2H, J = 9 Hz, HC(17)), 2.49 (t, J = 6.5 Hz, 2H, H2C(2)-succ), 2.33 (t, J = 10 Hz, 2H, H2C(2)-GABA), 2.31 (m, 4H, HC(4)), 2.17 (m, 2H, HC(16)), 2.12 (s, 6H, H3C(21)), 2.05 (m, 2H, HC(12)), 2.00 (m, 2H, HC(7)), 1.88 (m, 2H, HC(1)), 1.86 (m, 2H, HC(2)), 1.84 (quint, J = 10 Hz, 2H, H2C(3)-GABA), 1.69 (m, 2H, HC(15)), 1.65 (m, 2H, HC(16)), 1.61 (m, 2H, HC(11)), 1.58 (m, 2H, HC(2)), 1.57 (m, 2H, HC(7)), 1.49 (m, 2H, HC(8)), 1.48 (m, 2H, HC(11)), 1.45 (m, 2H, HC(12)), 1.21 (m, 2H, HC(15)), 1.15 (m, 2H, HC(14)), 1.12 (m, 2H, HC(1)), 1.02 (s, 6H, H3C(19)), 1.01 (m, 2H, HC(9)), 0.63 (s, 6H, H3C(18)); 13C NMR (125 MHz, CDCl3): δ = 209.5 (C(20)), 172.8 (CO-succ, ester)), 172.7 (CO(GABA)), 172.1 (CO-succ, amide), 169.0 (CO(Gly)), 139.6 and 139.5 (C(5)), 122.4 (C(6)), 74.6 and 74.0 (C(3)), 63.6 (C(17)), 56.8 (C(14)), 49.9 and 49.8 (C(9)), 43.9 (C(13)), 43.3 (C(Gly)), 39.0 (C(4)-GABA), 38.8 (C(12)), 38.0 (C(4)), 36.9 (C(1)), 36.6 (C(10)), 32.0 (C(2)-GABA)), 31.8 (C(8)), 31.7 (C(7)), 31.5 (C(21)), 30.9 (C(2)-succ), 29.8 (C(3)-succ), 27.7 (C(2)), 24.5 (C(3)-GABA), 24.4 (C(15)), 22.8 (C(16)), 21.0 (C(11)), 19.3 (C(19)), 13.2 (C(18)); 1H NMR (500 MHz, [D6]DMSO): δ = 8.16 (t, J = 6.0 Hz, 1H, NH(Gly)), 7.68 (t, J = 6.0 Hz, 1H, NH (GABA)), 5.33 (m, 2H, HC(6)), 4.45 (m, 2H, HC(3)), 3.61 (d, J = 5.5 Hz, 2H, H2C(Gly)), 3.06 (q, J = 6.5 Hz, 2H, H2C(4)-GABA), 2.56 (two t, J = 9.0 Hz, 2H, HC(17)), 2.49 (t, J = 6.5 Hz, 2H, H2C(3)-succ), 2.39 (t, J = 6.5 Hz, 2H, H2C(2)-succ), 2.25 (bt, J = 7.5 Hz, 6H, H2C(4) and H2C(2)-GABA), 2.05 (s, 6H, H3C(21)), 2.02 (m, 2H, HC(16)), 2.00 (m, 2H, HC(12)), 1.94 (m, 2H, HC(7)), 1.84 (m, 2H, HC(1)), 1.75 (m, 2H, HC(2)), 1.63 (quint, J = 7.0 Hz, 2H, H2C(3)-GABA), 1.61 (m, 2H, HC(15)), 1.56 (m, 4H, HC(11), HC(16)), 1.54 (m, 4H, HC(2), HC(7)), 1.42 (m, 2H, HC(12)), 1.40 (m, 4H, HC(8), HC(11)), 1.14 (m, 4H, HC(14), HC(15), 1.06 (m, 2H, HC(1)), 0.96 (m, 2H, HC(9)), 0.96 (s, 6H, H3C(19)), 0.52 (s, 6H, H3C(18)); 13C NMR (125 MHz, [D6]DMSO): δ = 208.8 (C(20)), 172.2 (CO-succ, ester; CO-GABA), 171.5 (CO-succ, amide), 169.1 (CO(Gly), 139.7 and 139.6 (C(5)), 122.1 and 122.0 (C(6)), 73.4 and 73.3 (C(3)), 62.7 (C(17)), 56.1 (C(14)), 49.4 (C(9)), 43.4 (C(13)), 42.3 (C(Gly)), 38.0 (C(12)), 37.9 (C(4)-GABA), 37.8 and 37.7 (C(4)), 36.6 (C(1)), 36.2 (C(10)), 31.4 (C(8) and C(7)), 31.3 (C(21)), 31.2 (C(2)-GABA), 29.9 (C(2)-succ), 29.4 (C(3)-succ), 27.4 (C(2)), 24.6 (C(3)-GABA), 24.1 (C(15)), 22.4 (C(16)), 20.7 (C(11)), 19.1 (C(19)), 13.0 (C(18)); HRMS (ESI-TOF) (m/z): calcd for C52H77N2O8 (M + H)+: 857.5674, found: 857.5664.
Rotaxane 4a. A solution of p-xylenediamine (135 mg, 0.988 mmol) in anh. CHCl3 (42 mL) and a separate solution of isophthaloyl dichloride (200 mg, 0.988 mmol) in anh. CHCl3 (42 mL) were simultaneously added dropwise over 4 h to a stirred solution of the thread 3a (85 mg, 0.066 mmol) and dry TEA (200 μL, 1.403 mmol) in dry CHCl3 (25 mL) under Ar. After the addition, the reaction mixture was stirred overnight at ambient temperature. The mixture was filtered through Celite, evaporated to dryness and chromatographed on silica gel by FCC. The pure unconsumed thread 3a (68 mg, 80%) was eluted with PhMe/EtOAc/MeOH 10
:
10
:
3 while the mixture of rotaxane and free macrocycle (14 mg) was eluted with PhMe/EtOAc/MeOH 10
:
10
:
7 and further separated by size exclusion chromatography on Sephadex G-25. Sephadex G-25 (10 g) was suspended in CHCl3 (stabilized with 1% EtOH), left to swell overnight and the solution of the mixture of 4a and free macrocycle (14 mg) in CHCl3/MeOH (95
:
5, 0.5 mL) was applied to the column. Elution with CHCl3, precipitation from highly concentrated CHCl3 solution with MeOH and subsequent drying under vacuum afforded the rotaxane 4a (9 mg, 8%) as a brown powder. Rf = 0.35 (PhMe/EtOAc/MeOH 5
:
5
:
1); UV/Vis (CHCl3): λmax(ε) = 428 nm (6.15 × 103 mol−1 dm3 cm−1); IR:
= 3428, 2951, 1726, 1141 cm−1. The 1H NMR spectrum of 4a in CDCl3 at a concentration of 15 mg mL−1 was obtained with low resolution due to the high level of self-aggregation. Dilution to 2.5 mg mL−1 provided an acceptable 1H NMR spectrum with slightly broadened signals (data given below) and a 13C NMR spectrum with a low signal-to-noise ratio (see S10;† data not given); 1H NMR (500 MHz, CDCl3): δ = 8.41 (s, 2H, M4), 8.13 (d, J = 8.0 Hz, 4H, M2), 7.60 (m, 6H, M1 and NH(M)), 7.26 (s, 8H, M6), 6.85 (br s, 1H, NH(GABA)), 5.90 (br s, 1H, NH(Gly)), 5.30 (m, 1H, HC(6)), 4.56 (s, 8H, H2C(M)), 4.51 (m, 1H, HC(3)), 4.27 (s, 4H, H2C(pyrr)), 3.06 (br s, 2H, H2C(4′)), 2.94 (br s, 2H, H2C(4′′)), 2.83 (br s, 2H, H2C(Gly)), 2.54 (t, 1H, J = 8.0 Hz, HC(17)), 2.25 (m, 2H, H2C(2′′)), 2.17 (m, 2H, H2C(2′), 1.91 (m, 2H, H2C(3′′)), 2.13 (s, 3H, H3C(21)), 1.66 (m, 2H, H2C(3′)), 0.97 (s, 3H, H3C(19)), 0.63 (s, 3H, H3C(18)); 1H NMR (500 MHz, [D6]DMSO + 1 drop of CDCl3): δ = 8.84 (m, 4H, NH(M)), 8.31 (s, 2H, M4)), 7.96 (m, 2H, NH(Gly) and NH(GABA)), 7.91 (d, J = 8.0 Hz, 4H, M2), 7.40 (br t, J = 8.0 Hz, 2H, M1), 7.24 (s, 8H, M6), 5.23 (m, 1H, HC(6)), 4.48 (m, 4H, H2C(M)), 4.32 (m, 4H, H2C(M)), 4.29 (m, 1H, HC(3)), 3.66 (s, 2H, H2C(Gly)), 3.38 (s, 4H, H2C(pyrr)), 2.92 (br s, 2H, H2C(4′)), 2.05 (m, 2H, H2C(4′′)), 2.03 (m, 2H, H2C(2′)), 1.88 (m, 2H, H2C(2′′)), 1.48 (m, 2H, H2C(3′)), 1.45 (m, 2H, H2C(3′′)), 0.91 (m, 1H, HC(9)), 0.86 (s, 3H, H3C(19)), 0.50 (s, 3H, H3C(18)); 13C NMR (125 MHz, [D6]DMSO + 1 drop of CDCl3): δ = 208.6 (C(20)), 173.6 (CO(1′′)), 172.1 (CO(1′)), 169.3 (CO(Gly)), 165.7 (CO(M)), Cfull (155.1, 146.8, 146.1, 145.7, 145.5, 145.3, 144.7, 144.1, 142.5, 142.1, 141.9, 141.6, 141.4, 139.4, 135.7), 139.5 (C(5)), 138.1 (C(6)-M), 134.5 (C(3)-M), 130.1 (C(2)-M), 128.8 (C(5)-M), 128.0 (C(1)-M), 126.1 (C(4)-M), 121.9 (C(6)), 73.1 (HC(3)), 69.9 (Cfull-sp3), 66.2 (C(pyrr)), 62.7 (C(17)), 56.1 (C(14)), 52.9 (C(4′′)), 49.3 (C(9)), 43.3 (CH2(M)), 43.2 (C(13)), 42.2 (C(Gly)), 38.0 (C(12)), 37.9 (C(4′)), 37.5 (C(4)), 36.5 (C(1)), 36.1 (C(10)), 32.6 (C(2′)), 31.4 (C(8), C(7)), 31.2 (C(21)), 27.2 (C(2)), 24.5 (C(3′′)), 24.1 (C(3′)), 22.3 (C(15)), 22.2 (C(16)), 20.6 (C(11)), 19.0 (C(19)), 13.0 (C(18)); HESI-Orbitrap MS (m/z): calcd for C125H80N7O9 (M + H)+: 1822.5939, found: 1822.6164.
Rotaxane 4b. A solution of p-xylenediamine (96 mg, 0.701 mmol) in CHCl3 (anhydrous, 15 mL) and a separate solution of isophthaloyl dichloride (143 mg, 0.701 mmol) in CHCl3 (15 mL) were added dropwise simultaneously for 4 h to a stirred solution of the thread 3b (40 mg, 0.047 mmol) and TEA (200 μL, 1.403 mmol) in dry CHCl3 (45 mL) under an atmosphere of Ar. After the addition, the reaction mixture was stirred overnight at ambient temperature. The mixture was filtered through Celite, evaporated to dryness and chromatographed on SiO2 by FCC. The unreacted thread (34 mg, 85%) was eluted with PhMe/EtOAc/MeOH 10
:
10
:
1 while rotaxane 4b was eluted with PhMe/EtOAc/MeOH 10
:
10
:
5 and precipitated from DCM solution with n-hexane. Yield: 8.2 mg (13%); Rf = 0.49 (PhMe/EtOAc/MeOH 5
:
5
:
1); IR:
= 3481, 3439, 3308, 3062, 2942, 2851, 1729, 1701, 1647, 1622, 1535, 1270, 1172, 737 cm−1; 1H NMR (500 MHz, CDCl3): δ = 8.33 (s, 2H, M4), 8.14 (dd, J = 1.0 and 8.0 Hz, 4H, M2), 7.58 (t, J = 8.0 Hz, 2H, M1), 7.52 (m, 4H, NH(M)), 7.22 (s, 8H, M6), 7.10 (br t, J = 5.5 Hz, 1H, NH(GABA)), 5.95 (br t, X part of ABX system, J = 4.0 Hz, 1H, NH(Gly)), 5.29 (m, 1H, HC(6)), 5.21 (m, 1H, HC(6)), 4.64 (m, 4H, H2C(M)), 4.45 (m, 4H, H2C(M)), 4.46 (m, 1H, HC(3)), 4.42 (m, 1H, HC(3)), 3.06 (q, J = 6.5 Hz, 2H, H2C(4)-GABA), 2.89 and 2.85 (two AB quartets, AB part of ABX system, J(AB) = 17 Hz, J(AX, BX) = 4.0 Hz, 2H, H2C(Gly)), 2.53 (t, J = 9.0 Hz, 2H, HC(17)), 2.32 (br t, J = 6.5 Hz, 2H, H2C(3)-succ), 2.19 (m, 6H, HC(4) and HC(16)), 2.14 (m, 2H, H2C(2)-GABA), 2.13 (s, 3H, H3C(21)), 2.12 (s, 3H, H3C(21)), 2.09 (br t, J = 6.5 Hz, 2H, H2C(2)-succ), 2.06 (m, 2H, HC(12)), 1.98 (m, 2H, HC(7)), 1.79 (m, 2H, HC(1)), 1.70 (m, 2H, HC(2)), 1.67 (m, 2H, HC(15)), 1.66 (m, 4H, HC(16) and H2C(3)-GABA), 1.59 (m, 2H, HC(11)), 1.55 (m, 2H, HC(7)), 1.49 (m, 2H, HC(2)), 1.46 (m, 2H, HC(11)), 1.45 (m, 4H, HC(8) and HC(12)), 1.23 (m, 2H, HC(15)), 1.15 (m, 2H, HC(14)), 0.99 (m, 2H, HC(1)), 0.97 (m, 2H, HC(9)), 0.95 (s, 6H, H3C(19)), 0.62 (s, 6H, H3C(18)); 13C NMR (125 MHz, CDCl3): δ = 209.6 (C(20)), 172.8 (CO-succ, ester), 172.5 (CO-GABA), 172.2 (CO-succ, amide), 169.1 (CO-Gly), 166.5 (CO-M), 139.2, 139.0 (C(5)), 137.7 (C(6)-M), 133.9 (C(3)-M), 131.5 (C(2)-M), 129.2 (C(1)-M), 129.1(C(5)-M), 124.0 (C(4)-M), 122.7 and 122.6 (C(6)), 74.8 and 74.4 (C(3)), 63.6 (C(17)), 56.8 (C(14)), 49.8 and 49.7 (C(9)), 44.1 (H2C-M), 44.0 (C(13)), 42.4 (C(Gly)), 39.2 (C(4)-GABA), 38.7 (C(12)), 37.9 (C(4)), 36.8 and 36.7 (C(1)), 36.5 (C(10)), 31.7 (C(7), C(8), C(20)), 31.5 (C(2)-GABA), 30.1 (C(2)-succ), 29.7 (C(3)-succ), 27.7 and 27.6 (C(2)), 24.5 (C(15)), 24.3 (C(3)-GABA), 22.8 (C(16)), 21.0 (C(11)), 19.2 (C(19)), 13.2 (C(18)); the unthreading of the rotaxane 4b started immediately after dissolution in [D6]DMSO, so the corresponding 1H and 13C chemical shifts were determined by comparative analysis of the NMR data of the sample, pure thread 3b and pure M. 1H NMR (500 MHz, [D6]DMSO): δ = 8.60 (br q, J = 5.0 Hz, 4H, NH(M)), 8.15 (s, 2H, M4), 7.94 (dd, J = 1.5 and 8.0 Hz, 4H, M2), 7.86 (br t, J = 5.5 Hz, 1H, NH(Gly)), 7.64 (br t, J = 5.5 Hz, 1H, NH(GABA)), 7.54 (t, J = 7.5 Hz, 2H, M1), 7.16 (s, 8H, M6), 5.16 (m, 1H, HC(6)), 5.12 (m, 1H, HC(6)), 4.39 (m, 4H, H2C(M)), 4.32 (m, 4H, H2C(M)), 4.12 (m, 1H, HC(3)), 3.99 (m, 1H, HC(3)), 3.38 (J = 5.5 Hz, 2H, H2C(Gly)), 2.58 (q, J = 6.0 Hz, 2H, H2C(4)-GABA), 2.54 (br t, J = 9.0 Hz, 2H, HC(17)), 2.10 (m, 4H, H2C(2)-succ and H2C(3)-succ), 2.04 (s, 3H, H3C(21)), 1.99 (m, 2H, HC(16)), 1.95 (m, 2H, HC(12)), 1.86 (m, 2H, HC(7)), 1.76 (m, 4H, HC(4)), 1.73 (m, 2H, H2C(2)-GABA), 1.62 (m, 2H, HC(1)), 1.57 (m, 2H, HC(15)), 1.54 (m, 2H, HC(16)), 1.48 (m, 2H, HC(11)), 1.45 (m, 2H, HC(7)), 1.38 (m, 2H, HC(12)), 1.31 (m, 4H, HC(8) and HC(12)), 1.22 (m, 2H, HC(2)), 1.18 (m, 2H, H2C(3)-GABA), 1.10 (m, 4H, HC(14) and HC(15)), 0.99 (m, 2H, HC(2)), 0.82 (m, 2H, HC(9)), 0.81 (s, 3H, H3C(19)), 0.78 (m, 2H, HC(1)), 0.75 (s, 3H, H3C(19)), 0.49 (s, 6H, H3C(18)); 13C NMR (125 MHz, [D6]DMSO): δ = 208.9 (C(20)), 172.3 (CO(GABA)), 172.1 (CO-succ, ester), 171.9 (CO-succ, amide), 169.0 (CO(Gly)), 166.0 (CO-M), 139.5 (C(5)), 137.7 (C(6)-M), 134.6 (C(3)-M), 130.1 (C(2)-M), 128.7 (C(1)-M), 128.5 (C(5)-M), 126.1 (C(4)-M), 121.8, 121.7 (C(6)), 73.3 and 73.0 (C(3)), 62.7 (C(17)), 56.1 (C(14)), 49.3 (C(9)), 43.2 (H2C-M), 42.8 (C(13)), 42.1 (C(Gly)), 38.0 (C(12)), 37.8 (C(4)-GABA), 37.3 and 37.1 (C(4)), 36.5 (C(1)), 36.1 (C(10)), 31.4 (C(8) and (C(7)), 31.3 (C(21)), 31.0 (C(2)-GABA), 29.7 (C(2)-succ), 28.9 (C(3)-succ), 27.0 and 26.8 (C(2)), 24.3 (C(3)-GABA), 24.1 (C(15)), 22.4 (C(16)), 20.6 (C(11)), 19.0 and 18.9 (C(19)), 13.0 (C(18)). HRMS (ESI-TOF) (m/z): calcd for C84H105N6O12 (M + H)+: 1389.7785, found: 1389.7765.
3. Results and discussion
Target compounds were prepared according to Scheme 1 considering the observation that labile hydrogen bonding in rotaxanes allows solvent-induced molecular motion.9a,b Hereof, the presence of the GABA-Gly-GABA diamide fragment and suitable bulky terminal subunits in the previously synthesized, quite soluble fullerene-peptide-steroid triad 3a19 generated an optimal thread for the unsymmetrical, fullerene-steroid-stoppered molecular shuttle. By analogy, a Succ-Gly-GABA component as the molecular recognition core as well as pregnenolone subunits as endpoints were used to design the thread 3b for a steroid-stoppered, interlocked system. The new compound 3b was prepared analogous to 3a, following the DCC/HOBT assisted amidation of the amine-TFA salt 1 using pregnenolone hemisuccinate 2b in the presence of triethylamine in dichloromethane. Following the Leigh procedure,22 isophthaloyl dichloride and p-xylylenediamine were simultaneously added to a solution of the corresponding threads 3. Their diamide GABA-Gly-GABA and GABA-Gly-Succ subunits, containing two amide carbonyl functions in the 1,4-position, templated the macrocycle clipping around the axes, providing rotaxanes 4. The careful flash column chromatography (FCC) of the crude, unsymmetrically stoppered, non-covalent hybrid on SiO2 gave a mixture of rotaxane 4a and the free tetraamide macrocycle (M). The conditions for their purification by chromatography on silica gel were not found, since both compounds showed equal Rf values in many applied eluents (numerous binary and ternary mixtures of CHCl3, PhMe, EtOAc and MeOH). The favorable π–π interactions between the free macrocycle and the outer side of the fullerene subunit might be a reason for unsuccessful separation on the silica column. Nevertheless, pure rotaxane 4a was obtained by gel filtration on Sephadex G-25 in CHCl3, subsequent precipitation from CHCl3 solution with MeOH and drying in a vacuum oven for two days (45 °C, 15 mbar). In the NMR spectrum in CDCl3, signals corresponding to poorly soluble free M did not appear, while in [D6]DMSO diluted with one drop of CDCl3 (1 drop of CDCl3 was necessary to completely dissolve the mixture 4a/M) they were clearly visible, providing a way to check unambiguously the purity of rotaxane 4a.
 |
| Scheme 1 Synthesis of fullerosteroidal and disteroidal rotaxanes 4 with a tetraamide macrocycle. | |
In contrast, a careful FCC of the disteroidal rotaxane 4b proceeded smoothly, providing a pure target compound. Although performing hydrogen bond-templated interlocking afforded rotaxanes in quite a moderate yield (Scheme 1), their precursors were successfully recovered (80% of 3a and 85% of 3b) and reused. In addition to the structural requirements, the achieved yield might be attributed to the poor solubility of thread precursors and the unthreading process in the presence of polar medium during column chromatography.
The structures of all compounds were determined from IR, NMR (1H, 13C, COSY, HSQC and HMBC) and HRMS spectra supported also by the UV/Vis data of the fullerene containing thread 3a and rotaxane 4a. The presence of a steroidal moiety in the rotaxane 4a significantly improved its solubility in chloroform, but concentrations higher than 1.4 mM (2.5 mg mL−1, used for 1H NMR recording) led to solution jellification, indicating the compounds self-aggregate. Also, in the 13C NMR spectrum in CDCl3 no response for thread carbonyls was observed, while [D6]DMSO (with addition of one drop of CDCl3) enabled assignment of all carbons. Comparison of the 1H NMR spectra of rotaxanes 4 and the corresponding threads 3 in CDCl3 (Fig. 1) affirmed the interlocked architecture anchored by hydrogen bonds between the four macrocyclic amide H and the two amide carbonyl oxygen atoms. The location of the macrocycle on the diamide station (GABA-Gly-GABA in 4a and Succ-Gly-GABA in 4b, Scheme 2) was clearly determined from the chemical shift differences of the pyrrolidine, amide and steroid protons in rotaxanes and corresponding threads (Fig. 1, Tables S1 and S2 in ESI†).
 |
| Fig. 1 Partial 1H NMR spectra (500 MHz, CDCl3) of (a) thread 3a, (b) rotaxane 4a, (c) thread 3b, and (d) rotaxane 4b. Labels correspond to those indicated in Scheme 2. | |
 |
| Scheme 2 Solvent-induced translocation of the macrocyclic ring (M) with the proposed structures of co-conformers followed by partial and complete dissociation of rotaxanes 4a and 4b, respectively. | |
Thus, as a result of the anisotropic effect of the macrocycle aromatic rings, both methylene and amide Gly protons of rotaxanes 4 are significantly upfield shifted in comparison to threads 3 (Δδ ∼ 1 ppm for CH2; 1.0 and 0.5 ppm for NH of 4a and 4b, respectively). Somewhat weakened shielding of protons near the Gly N-side resulted in lesser shifting of ∼0.4 ppm (GABA 2′′, 3′′ and Succ 2, 3 in 4a and 4b, respectively, Tables S1 and S2†). The smallest difference (up to 0.2 ppm) in the chemical shifts of the remaining signals belonging to the pyrrolidine ring, the rest of GABA and characteristic steroidal protons (H-3, H-6 and the angular methyl groups) confirmed the position of the macrocycle around the diamide binding site, as shown in Scheme 2. In addition, diverse shielding of the steroidal stoppers by the macrocycle aromatic rings led to the splitting of the olefinic H-6 signal (Fig. 1c and d and 2d and e). Samples of both rotaxanes were kept in CDCl3 solution for a week at room temperature with no NMR evidence of macrocycle release.
 |
| Fig. 2 Partial 1H NMR spectra (500 MHz, [D6]DMSO) of (a) thread 3a, (b) rotaxane 4a immediately after dissolution, (c) rotaxane 4a after 24 h, (d) thread 3b, (e) rotaxane 4b immediately after dissolution, (f) rotaxane 4b after 24 h, and (g) free macrocycle M. For labels, see Scheme 2. | |
Exposing these compounds to solvents possessing strong H-bond accepting properties, such as DMSO, induced macrocycle shuttling along the corresponding threads. Thanks to the fullerene stopper and the resulting π–π interactions with four aromatic rings of the wheel,9a,b the mechanically interlocked architecture of rotaxane 4a was mainly preserved, while under the same conditions, the disteroidal analogue 4b appeared less stable. A striking upfield shift of the pyrrolidinic signal of rotaxane 4a in DMSO (δ 3.38 ppm, Fig. 2b) in comparison to the same structure in CDCl3 (δ 4.27 ppm, Fig. 1b), as well as to thread 3a in DMSO (δ 4.45 ppm, Fig. 2a), was noticed, indicating wheel displacement to the fullerene proximity (see also Table S1†). Also, the resonances associated with methylene protons belonging to the GABA moiety adjacent to the pyrrolidine (2′′–4′′, Scheme 2) followed the same trend. Observed shifts (reducing the distance from the C60 moiety) additionally supported the macrocycle position (Fig. 2a–c and Table S1†) and formation of the stacked co-conformer of 4a in [D6]DMSO (Scheme 2). The resonance of macrocycle benzylic protons (CH2(M)) appeared as a broad singlet only in rotaxane 4a in CDCl3 (Fig. 1b), indicating their fast exchange. The signal splitting (two multiplets for AB part of ABX system) in rotaxane 4b in CDCl3 (Fig. 1d) and in both rotaxanes in [D6]DMSO (Fig. 2b and e) supports its reduced symmetry and indicates slow rotation of the macrocycle around the thread on the NMR time scale at room temperature.
In order to monitor shuttling processes in more detail, 1H NMR titration of rotaxane 4a in CDCl3 with [D6]DMSO was performed (Fig. 3). As can be seen, the first addition of [D6]DMSO (1%, Fig. 3b) led to complete solvation of the GABA-amide proton provoking its strong downfield shift from δ 6.85 to 7.79 ppm, practically to the value observed for the free thread. In such conditions the chemical shift of the Gly amide proton was not changed, probably due to sterically hindered solvent access. A gradual increase of the medium polarity induced the progressive breakage of the wheel-thread H-bonds and the solvation of the macrocycle amide protons (NH(M)), leading to their deshielding of almost 1 ppm – from δ 7.79 ppm in the presence of 1% [D6]DMSO to δ 8.76 ppm in the mixture containing 52% [D6]DMSO (Fig. 3b–i). After addition of 15% [D6]DMSO (Fig. 3f), several changes in the 1H NMR spectrum were observed, all together suggesting the parallel existence of macrocycle conformational changes and solvation of Gly-NH. Thus, a broad singlet of CH2(M) changed to two multiplets and the CHPyrr2 singlet was slightly shifted upfield (Δδ ∼ 0.2 ppm), while Gly-NH proton was shifted downfield (Δδ ∼ 0.8 ppm). A low level of unthreading was first detected with 42% [D6]DMSO, resulting in the appearance of weak signals corresponding to the free thread as well as the macrocycle (Fig. 3h). Increasing the medium polarity to 52% [D6]DMSO led to the upfield shift of the pyrrolidine singlet to δ < 3.70 (overlapped with water signal) and further deshielding of both amide protons (Fig. 3i), indicating macrocycle motion preferentially toward the fullerene stopper and formation of the stable stacked co-conformer (Scheme 2). No significant dissociation of fullerosteroidal rotaxane 4a in [D6]DMSO was detected over 24 h (2 and 10% after 1 and 24 h, respectively), while after 7 days in the same solvent 25% of the mechanically interlocked system underwent unthreading. In contrast, rigid but not sufficiently bulky pregnenolone units in the steroidal rotaxane 4b allowed the macrocycle to slip over them, enabling a much faster dissociation process (4% after 1 h, 100% after 24 h). In such a case, many intermediate steroid co-conformers (Scheme 2) might be suggested. Due to the fast unthreading in a polar solvent at room temperature, in such conditions the supramolecular assembly 4b could be considered as a pseudorotaxane.
 |
| Fig. 3 1H NMR monitoring of macrocycle translocation in the rotaxane 4a in CDCl3 (a) induced by [D6]DMSO (b: 1%, c: 2%, d: 4%, e: 8%, f: 15%, g: 30%, h: 42%, i: 52%). | |
Previously reported results confirming organogelation behaviour of the fullerene–cholesterol conjugate23 and cholesterol-stoppered rotaxane,18 prompted us to examine the self-organization properties of synthesized rotaxanes 4. To that purpose, the solubility in individual solvents (CHCl3, PhMe, THF, dioxane and EtOAc) and their binary mixtures was checked, while the form of the aggregates obtained after solvent evaporation was followed by scanning electron microscopy (SEM). The presence of the fullerene moiety in rotaxane 4a reduced its solubility, affording a clear solution only in chloroform. The jellification noticed at higher concentration (1.4 mM) during NMR recording indicated reorganization at the supramolecular level. The direct evidence for the tendency for self-ordering was obtained by SEM, the image of which revealed the association of quite uniform, round structures to micrometer-sized elongated particles (Fig. 4a). As expected, both disteroidal structures (thread 3b and rotaxane 4b) were easily soluble in all media used. However, unlike compound 3b, after few minutes at room temperature a clear 1 mM solution of 4b in PhMe/EtOAc 1
:
1 mixture was transformed into transparent gel, stable up to 50 °C (Fig. 4c). Prolonged drying of a drop of 0.1 mM solution allowed organization of 4b into a branched film-like network (Fig. 4b) or into irregularly shaped large microstructures in the case of 3b (Fig. S1 in ESI†), confirming their different affinities for aggregation. The observed difference in the gelation ability between the disteroidal thread 3b and the corresponding rotaxane 4b strongly indicates that hydrogen bonding interactions of the mechanically interlocked structure plays an important role in self-assembly, thereby controlling the gelation process.
 |
| Fig. 4 SEM images of (a) rotaxane 4a (0.1 mM in CHCl3), (b) rotaxane 4b (0.1 mM in PhMe/EtOAc = 1 : 1), and (c) photo of 4b gel (1 mM in PhMe/EtOAc = 1 : 1). | |
4. Conclusions
Two novel hydrogen-bond assembled fullerosteroidal and disteroidal[2]rotaxanes are synthesized by GABA-Gly-GABA and Succ-Gly-GABA templated macrocycle clipping over the axles containing corresponding stoppers. 1H NMR spectroscopy was used to investigate solvent-induced macrocycle movement along the threads, as well as the stability of the interlocked architectures, expressed as the affinity for unthreading. The target compounds have proven to be stable in a non polar environment (CDCl3) regardless of the stopper nature, occupying the peptide co-conformation with the macrocycle anchored by the axle–wheel interamide H-bonds. In the case of the fullerene-stoppered rotaxane 4a, the solvation of amide groups by [D6]DMSO caused macrocycle translocation toward the fullerene and formation of a quite stable stacked co-conformer, stabilized by wheel-stopper π–π interactions. In contrast, rigid steroidal moieties turned out to be insufficiently bulky to protect the macrocycle from slipping, leading to practically quantitative unthreading of the disteroidal rotaxane 4b after 24 h. In addition, both rotaxanes expressed a strong tendency to self-order, while a rotaxane with terminal pregnenolone units also showed organogelation behaviour. Based on their dynamic properties, these novel supramolecular architectures could be of use for designing new interlocked systems with controlled dissociation in polar medium.
Acknowledgements
This work has been supported by the Ministry of Education, Science and Technological Development of Serbia (Project No. 172002).
Notes and references
-
(a) M. Xue, Y. Yang, X. Chi, X. Yan and F. Huang, Chem. Rev., 2015, 115, 7398 CrossRef CAS PubMed;
(b) H. Tian and Q.-C. Wang, Chem. Soc. Rev., 2006, 35, 361 RSC;
(c) X. Ma and H. Tian, Chem. Soc. Rev., 2010, 39, 70 RSC;
(d) F. Vögtle, T. Dünwald and T. Schmidt, Acc. Chem. Res., 1996, 29, 451 CrossRef.
- A. Mateo-Alonso, G. Fioravanti, M. Marcaccio, F. Paolucci, D. C. Jagesar, A. M. Brouwer and M. Prato, Org. Lett., 2006, 8, 5173–5176 CrossRef CAS PubMed.
-
(a) A. Fernandes, A. Viterisi, F. Coutrot, S. Potok, D. A. Leigh, V. Aucagne and S. Papot, Angew. Chem., Int. Ed., 2009, 48, 6443 CrossRef CAS PubMed;
(b) R. Barat, T. Legigan, I. Tranoy-Opalinski, B. Renoux, E. Péraudeau, J. Clarhaut, P. Poinot, A. E. Fernandes, V. Aucagne, D. A. Leigh and S. Papot, Chem. Sci., 2015, 6, 2608 RSC.
- R. Hernandez, H.-R. Tseng, J. W. Wong, J. F. Stoddart and J. I. Zink, J. Am. Chem. Soc., 2004, 126, 3370 CrossRef CAS PubMed.
- J. Zhu, M. McFarland-Mancini, A. F. Drew and D. B. Smithrud, Bioorg. Med. Chem. Lett., 2009, 19, 520 CrossRef CAS PubMed.
- J. R. Johnson, N. Fu, E. Arunkumar, W. M. Leevy, S. T. Gammon, D. Piwnica-Worms and B. D. Smith, Angew. Chem., Int. Ed., 2007, 46, 5528 CrossRef CAS PubMed.
-
(a) A. Rescifina, C. Zagni, D. Iannazzo and P. Merino, Curr. Org. Chem., 2009, 13, 448 CrossRef CAS;
(b) E. R. Kay, D. A. Leigh and F. Zerbetto, Angew. Chem., Int. Ed., 2007, 46, 72 CrossRef CAS PubMed;
(c) C. J. Bruns and J. F. Stoddart, Acc. Chem. Res., 2014, 47, 2186 CrossRef CAS PubMed;
(d) T. D. Nguyen, H.-R. Tseng, P. C. Celestre, A. H. Flood, Y. Liu, J. F. Stoddart and J. I. Zink, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10029 CrossRef CAS PubMed.
-
(a) D. B. Amabilino and J. F. Stoddart, Chem. Rev., 1995, 95, 2725 CrossRef CAS;
(b) D.-H. Qu and H. Tian, Chem. Sci., 2011, 2, 1011 RSC;
(c) G. Barin, R. S. Forgan and J. F. Stoddart, Proc. R. Soc. London, Ser. A, 2012, 468, 2849 CrossRef CAS PubMed.
-
(a) A. Mateo-Alonso, D. M. Guldi, F. Paolucci and M. Prato, Angew. Chem., Int. Ed., 2007, 46, 8120 CrossRef CAS PubMed;
(b) A. Mateo-Alonso, Top. Curr. Chem., Fullerene-stoppered bistable rotaxanes, Springer Press, Berlin, Heidelberg, 2014, 348, pp. 127–137 Search PubMed;
(c) A. S. Lane, D. A. Leigh and A. Murphy, J. Am. Chem. Soc., 1997, 119, 11092 CrossRef CAS.
- J. J. Gassensmith, E. Arunkumar, L. Barr, J. M. Baumes, K. M. DiVittorio, J. R. Johnson, B. C. Noll and B. D. Smith, J. Am. Chem. Soc., 2007, 129, 15054 CrossRef CAS PubMed.
- E. A. Neal and S. M. Goldup, Chem. Commun., 2014, 50, 5128–5142 RSC.
-
(a) H. R. Tseng, S. A. Vignon and J. F. Stoddart, Angew. Chem., Int. Ed., 2003, 42, 1491 CrossRef CAS PubMed;
(b) A. Altieri, F. G. Gatti, E. R. Kay, D. A. Leigh, D. Martel, F. Paolucci, A. M. Z. Slawin and J. K. Y. Wong, J. Am. Chem. Soc., 2003, 125, 8644 CrossRef CAS PubMed;
(c) F. Durola and J. P. Sauvage, Angew. Chem., Int. Ed., 2007, 46, 3537 CrossRef CAS PubMed.
-
(a) A. M. Elizarov, S. H. Chiu and J. F. Stoddart, J. Org. Chem., 2002, 67, 9175 CrossRef CAS PubMed;
(b) C. M. Keaveney and D. A. Leigh, Angew. Chem., Int. Ed., 2004, 43, 1222 CrossRef CAS PubMed.
-
(a) A. M. Brouwer, C. Frochot, F. G. Gatti, D. A. Leigh, L. Mottier, F. Paolucci, S. Roffia and G. W. H. Wurpel, Science, 2001, 291, 2124 CrossRef CAS PubMed;
(b) V. Balzani, M. Clemente-León, A. Credi, B. Ferrer, M. Venturi, A. H. Flood and J. F. Stoddart, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 1178 CrossRef CAS PubMed.
- G. Bottari, F. Dehez, D. A. Leigh, P. J. Nash, E. M. Pérez, J. K. Y. Wong and F. Zerbetto, Angew. Chem., Int. Ed., 2003, 42, 5886 CrossRef CAS PubMed.
- T. Da Ros, D. M. Guldi, A. F. Morales, D. A. Leigh, M. Prato and R. Turco, Org. Lett., 2003, 5, 689 CrossRef CAS PubMed.
- A. Mateo-Alonso and M. Prato, Tetrahedron, 2006, 62, 2003 CrossRef CAS.
- Y.-L. Zhao, I. Aprahamian, A. Trabolsi, N. Erina and J. F. Stoddart, J. Am. Chem. Soc., 2008, 130, 6348 CrossRef CAS PubMed.
- M. S. Bjelaković, T. J. Kop, M. Vlajić, J. Đorđević and D. R. Milić, Tetrahedron, 2014, 70, 8564 CrossRef.
- M. S. Bjelaković, D. M. Godjevac and D. R. Milić, Carbon, 2007, 45, 2260 CrossRef.
- J. D. Cocker, J. Elks, P. J. May, F. A. Nice, G. H. Phillipps and W. F. Wall, J. Med. Chem., 1965, 8, 417 CrossRef CAS PubMed.
- D. A. Leigh, A. Murphy, J. P. Smart and A. M. Z. Slawin, Angew. Chem., Int. Ed. Engl., 1997, 36, 728 CrossRef CAS.
- T. Ishi-i, Y. Ono and S. Shinkai, Chem. Lett., 2000, 29, 808 CrossRef.
Footnote |
† Electronic supplementary information (ESI) available: Tabular representation of the selected 1H NMR shifts, NMR and mass spectra of compounds 3 and 4. See DOI: 10.1039/c6ra03872g |
|
This journal is © The Royal Society of Chemistry 2016 |
Click here to see how this site uses Cookies. View our privacy policy here.